|Publication number||US20030067880 A1|
|Application number||US 10/179,582|
|Publication date||Apr 10, 2003|
|Filing date||Jun 24, 2002|
|Priority date||Oct 10, 2001|
|Also published as||EP1303111A2, EP1303111A3, US7289437|
|Publication number||10179582, 179582, US 2003/0067880 A1, US 2003/067880 A1, US 20030067880 A1, US 20030067880A1, US 2003067880 A1, US 2003067880A1, US-A1-20030067880, US-A1-2003067880, US2003/0067880A1, US2003/067880A1, US20030067880 A1, US20030067880A1, US2003067880 A1, US2003067880A1|
|Original Assignee||Girish Chiruvolu|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (5), Referenced by (39), Classifications (17), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
 This nonprovisional application claims priority based upon the following prior U.S. provisional patent application entitled: INFORMED DYNAMIC SHARED PATH PROTECTION, Provisional Application No. 60/328,087, filed Oct. 10, 2001, which is hereby incorporated by reference for all purposes.
 1. Technical Field of the Invention
 The present invention generally relates to optical networks. More particularly, and not by way of any limitation, the present invention is directed to a system and method for routing stability-based integrated traffic engineering (“RITE”) for a Generalized Multi-Protocol Label Switching (“GMPLS”) network.
 2. Description of Related Art
 GMPLS optical networks generally include a plurality of edge nodes, comprising MPLS routers, that have appropriate opto-electrical interfaces for converting incoming electrical signals to optical signals (E/O interfaces) at an “ingress node” and for converting optical signals to electrical signals (O/E interfaces) at an “egress node”. Traffic is routed between an ingress and egress node pair via optical cross connects (“OXCs”) located throughout the network. OXCs are nodes that lack any sort of electrical capability; they simply route optical signals from one node to another. A light channel (“LC”), or Label Switched Path (“LSP”), is an optical communications channel from an ingress node to an egress node via one or more intermediate nodes, which may comprise OXCs as well as edge nodes, in some cases.
 High-priority (“HP”) and low-priority (“LP”) traffic trunks are characterized by the start and end times of their active resource utilization, also known as “lifetimes”. HP traffic trunks need absolute routing stability. In contrast, LP traffic trunks may be rerouted; that is, traffic may be transported between an ingress/egress node pair via an LC on which it was not originally mapped. Rerouting stability is a measure of how many times a traffic trunk is rerouted in its lifetime. Lifetimes of traffic trunks are measured in time units (e.g., seconds, minutes), and are typically drawn from a probabilistic distribution.
 A direct LC is an LC that is established between an ingress/egress node pair that constitutes only a single LC. In contrast, a multi-hop LSP constitutes one or more edge nodes as intermediate termination points of the LCs and may comprise multiple wavelengths. The traffic undergoes O/E/O conversion at the intermediate edge nodes. HP and LP requests are requests for bandwidth allocation and appropriate mapping on the LCs for HP and LP traffic trunks, respectively. Mapping could be on existing, or established, LCs or on new LCs that are established dynamically.
 Static provisioning of light channels in a GMPLS optical network is generally cumbersome and time-consuming and often leads to underutilization of resources if traffic demands within the network vary with time. In order to adapt to the changing traffic demands for optimal resource utilization, schemes that dynamically establish and teardown optical channels are highly desirable. However, this “make-and-break” of optical channels often results in the rerouting of existing traffic that can lead to better utilization of resources. Rerouting for individual traffic trunks can affect many relevant quality of service (“QoS”) measures, such as delay and jitter, and may lead to degradation of throughput. For example, Transmission Control Protocol (“TCP”) applications that do not have large buffers for reordering of the out-of-sequence packets may trigger retransmission of many packets. Delay and jitter are critical to real-time applications and also get affected as a result of rerouting of traffic trunks. In the case of non-packet-based Time Division Multiplex (“TDM”) signals, rerouting may translate into disruption of ongoing signals. Traffic Engineering (“TE”) procedures should address these issues in the context of dynamic wavelength assignment and traffic mapping in order to make optimal use of the optical resources.
 Currently, only general TE frameworks for GMPLS in optical networks have been proposed. No solutions currently exist that address the above-noted TE issues with respect to rerouting of traffic.
 One embodiment of a Routing Stability-Based Integrated Traffic Engineering (“RITE”) method described herein is directed toward a GMPLS/optical network in which the edge (i.e., ingress and egress) nodes comprise MPLS routers that are interconnected by an optical network that imposes wavelength continuity constraints across the end points (i.e., the MPLS routers). Light Channels (“LCs”), or Label Switched Paths (“LSPs”), are established on demand depending on the traffic loads at the MPLS routers. The goal of the RITE method is to increase utilization of existing/established LCs carrying maximum traffic, while conservatively establishing new LCs and minimizing the frequency with which traffic trunks are rerouted during their lifetime, which is defined as the time period between the establishment and subsequent tear down of the LC at the ingress node. LCs that are underutilized are torn down, thereby freeing up the relevant optical resources (e.g., wavelengths), and new LCs are established responsive to incoming traffic demands. Incoming traffic is classified as high priority (“HP”), which requires absolute routing stability, or low priority (“LP”), which can tolerate QoS degradation due to limited rerouting but subject to a limit on the frequency of the rerouting.
 In accordance with one embodiment, HP traffic trunks are mapped on to direct LCs that connect a designated pair of ingress and egress nodes and are rerouted only in the event of an LC/LSP teardown due to poor traffic utilization. LP traffic trunks are mapped on to direct LCs if available; otherwise, they are mapped on to LSPs that span multiple LCs with appropriate O/E and E/O conversions at the edge nodes serving as intermediate hops. The O-E-O conversions at the intermediate nodes may introduce packet delays for the traffic mapped on to multi-hop LSPs. Each such LP traffic trunk is associated with a rerouting timer that is set at the time of rerouting so as to prevent another rerouting of the trunk until the timer expires. The policy of the, RITE method described herein is to accommodate HP traffic on the existing LCs and reroute existing LP traffic trunks to multi-hop LCs as necessary to free up bandwidth on direct LCs for HP traffic.
 A more complete understanding of the present invention may be had by reference to the following Detailed Description when taken in conjunction with the accompanying drawings wherein:
FIG. 1 is a block diagram of a GMPLS/optical network in which features of one embodiment of a Routing-Stability Based Integrated Traffic Engineering (“RITE”) method may be implemented;
FIG. 2 depicts a Preferred Hops table of one embodiment of the RITE method of the present invention;
FIG. 3 depicts a Wavelengths Available table of one embodiment of the RITE method of the present invention;
FIGS. 4 and 5 depict flowcharts illustrating of the operation of one embodiment of the RITE method of the present invention;
FIG. 6 depicts a flowchart of a method of monitoring the utilization of each established LC of an optical network in accordance with one embodiment of the RITE method of the present invention; and
FIG. 7 depicts the results of a simple simulation performed as an evaluation of the RITE method described herein.
 In the drawings, like or similar elements are designated with identical reference numerals throughout the several views thereof, and the various elements depicted are not necessarily drawn to scale.
FIG. 1 depicts a simplified exemplary GMPLS optical network 100 comprising three MPLS, or edge, nodes 102 a, 102 b, and 102 c, and three optical cross-connects (“OXCs”) 104 a, 104 b, and 104 c. As previously noted, OXCs, such as the OXCs 104 a, 104 b, and 104 c, merely carry and switch light signals; they do not have O/E or E/O conversion capability. In contrast, the MPLS edge nodes (e.g., nodes 102 a, 102 b, and 102 c) do have such conversion capability. OXCs 104 a, 104 b, and 104 c are interconnected via links 106 a, 106 b, and 106 c. Node 102 a is connected to OXC 104 a via a link 108 a, node 102 b is connected to OXC 104 b via a link 108 b, and node 102 c is connected to OXC 104 c via a link 108 c.
 As illustrated in FIG. 1, requests for bandwidth are received by the edge nodes 104 a, 104 b, and 104 c. Each request is of a fixed size (e.g., 10 Mbits/second) and each of the LCs to be established carries a certain number of frequencies, so each LC can accommodate a fixed number (e.g., 10) of requests. Requests are classified into high priority (“HP”) and low priority (“LP”) requests according to the class of traffic the corresponding traffic trunks will be carrying. HP traffic requires absolute routing stability (i.e., no rerouting); in contrast, LP traffic can tolerate QoS degradation due to limited rerouting.
 A direct LC, or LSP, is one that comprises a direct optical connection between an ingress/egress node pair via one or more OXCs. In contrast, a multi-hop LC, or LSP, is one that constitutes more than one LC and hence comprises an optical connection between an ingress/egress node pair via one or more OXCs and one or more edge nodes other than the ingress and egress nodes. For example, referring to FIG. 1, assuming that a bandwidth request is received at the node 102 a to establish an LC between the node 102 a (designated the ingress node) and the node 102 b (designated the egress node), one example of a direct LC between the nodes is represented by a path 110. As illustrated in FIG. 1, the path 110 comprises the links 108 a, 106 a, and 108 b and includes the ingress node (node 102 a), OXC 104 a, OXC 104 b, and the egress node (node 102 b). It will be recognized that multiple direct LCs may exist for the same ingress/egress node pair. For example, a path 112 from the node 102 a to the node 102 b via the links 108 a, 106 c, 106 b, and 108 b, and including OXC 104 a, OXC 104 c, and OXC 104 b, also constitutes a direct LC for the ingress/egress node pair. In contrast, a path 114, which includes the links 108 a, 106 c, 108 c, 106 b, and 108 b, as well as node 102 c and OXCs 104 a, 104 c, and 104 b, represents a multi-hop LC for the ingress/egress node pair (nodes 102 a and 102 b).
 For purposes that will be described in greater detail below, each edge node 102 a, 102 b, and 102 c maintains two tables, including a Preferred Hops table 200 (FIG. 2) and a Wavelengths Available table 300 (FIG. 3). A description of each table, with specific reference to the network 100, follows.
 The Preferred Hops table 200 is a two-dimensional array/table indexed on the edge nodes only. The table 200 is updated whenever relevant Link State Advertisements (“LSAs”), such as LC establishment or over- or underutilization notifications are received. For example, when a new LC is established between the edge nodes i, j, where “i” designates the ingress node and “j” designates the egress node, then the corresponding entry (“preferred_hops[i] [j]”) of the Preferred Hops table 200 is incremented by one, indicating the availability of an additional LC between the nodes. If no LC exists between the edge nodes i, j, then the corresponding entry is set to zero. Whenever utilization of an LC crosses an upper threshold, the corresponding entry of the table is decremented by one upon receipt of the LSA. Similarly, when an LC is torn down, the corresponding entry of the table is decremented by one.
 Thus, each entry preferred_hops[i] [j] of the Preferred Hops table 200 indicates the number of direct LCs that exist between the ingress/egress node pair comprising the nodes i, j, that have spare capacity. For example, an entry 202 (“preferred_hops[node 102 a] [node 102 b]”) indicates the number of direct LCs between nodes 102 a and 102 b. For a multi-hop LSP, the Preferred Hops table 200 is consulted Dijkstra's algorithm is used to computer the shortest path between the given nodes. Specifically, as will be recognized by those of ordinary skill in the art, each of the links carries a “weight/metric” that can be chosen based on specific requirements. For example, if each of the links is characterized by distance in miles, then running Dijkstra's algorithm provides the sequence of nodes and the links to be followed to get from the source node to the destination node wherein the total route length is minimum. In the present embodiment, using the RITE method, the metric is selected to be the number of wavelengths/LCs available on a fiber link, resulting in a greater chance of finding a single wavelength, or LC, from one edge node (source) to another edge node (destination).
 The Wavelengths Available table 300 is also a two-dimensional array, but it is indexed by all the nodes, including OXCs as well as the edge nodes. Each entry of the table (“w_a[k] [l]”) indicates the number of wavelengths, or channels, available at the node on the link from node k to node l. For example, an entry 302 (“w_a[node 102 a] [OXC 104 a]”) indicates the number of available wavelengths on the link 108 a between node 102 a and OXC 104 a. It will be noted that no entries exist for node pairs between which there is no direct link (e.g., nodes 102 a and 102 b).
 The table 300 is updated whenever an LSA corresponding to an LC establishment or teardown is received. The Wavelengths Available table 300 is consulted in determining a route across the OXCs for the establishment of direct-LCs between a designated pair of edge nodes. The table 300 may be augmented with parameters such as distance or propagation time between any two nodes that are connected by direct links and appropriate transformations that can be done so as to run Dijkstra's algorithm to determine the shortest path, as described above.
FIG. 4 is a flowchart of the operation of one embodiment of the RITE method of the present invention. In step 400, responsive to receipt of a HP request for bandwidth for an HP traffic trunk between a designated ingress/egress node pair, a determination is made whether capacity exists on a direct LC between the designated ingress/egress node pair. If so, execution proceeds to step 402 in which the HP traffic trunk corresponding to the HP request is mapped to the existing direct LC. Otherwise, execution proceeds to step 404. In step 404, a determination is made whether a victim LP traffic trunk on the direct LC between the ingress/egress node pair is available for rerouting to another LC. To be eligible for rerouting, an LP traffic trunk must not have been rerouted within the predetermined preceding time period; in other words, its reroute timer must be expired. If an eligible victim LP traffic trunk is found, execution proceeds to step 406, in which the victim LP traffic trunk is rerouted to a multi-hop LC and the reroute timer for the LP traffic trunk is set, so that it cannot be re-rerouted until the timer expires (e.g., one hour). Execution then proceeds to step 408, in which the HP traffic trunk is mapped on to the direct LC on to which the victim LP traffic trunk was previously mapped.
 If in step 404, a victim LP was not found to be available, execution proceeds to step 410, in which a determination is made whether it is permissible for the HP traffic trunk corresponding to the HP request to be mapped on to a multi-hop LC. This will be the case if the HP class does not impose constraints that make the delay due to O-E-O conversions inherent in a multi-hop LC intolerable. It will be recognized that HP trunks mapped on to multi-hop LCs stand a greater risk of rerouting due to tear down of any one of the underlying direct LCs due to underutilization. Generally, it will be presumed that HP trunks must be mapped to direct LCs. If it is determined that it is not permissible for the HP traffic trunk corresponding to the received HP request to be mapped on to a multi-hop LC , as would generally be the case, execution proceeds to step 412, in which a new direct LC is created, and then to step 414, in which the HP traffic trunk is mapped on to the newly-created direct LC.
 Assuming, however, that it is not impermissible for the HP traffic trunk to be mapped on to a multi-hop LC, execution proceeds from step 410 to step 416, in which the existing multi-hop LCs between the ingress/egress node pair are checked to determine whether sufficient capacity exists on any one of them. If so, execution proceeds to step 418, in which the HP traffic trunk is mapped to the existing multi-hop LC. If not, execution proceeds to step 412 (described above).
 Requests for bandwidth for HP and LP class traffic trunks (respectively referred to as HP requests and LP requests) arrive at the ingress edge nodes. FIGS. 5 and 6 respectively depict flowcharts illustrating the mapping of HP and LP requests/trunks on to LCs in accordance with the RITE method of one embodiment. In particular, referring to FIG. 5, responsive to receipt of an LP request for bandwidth for an LP traffic trunk between a designated ingress/egress node pair, a determination is made whether capacity exists on a direct LC between the designated ingress/egress node pair. If so, execution proceeds to step 502 in which the HP traffic trunk corresponding to the HP request is mapped to the existing direct LC. Otherwise, execution proceeds to step 504. In step 504, a determination is made whether a multi-hop LC is available on to which to map the LC traffic trunk. In this step, the Preferred Hops table is consulted to compute a feasible path for a multi-hop LSP between the designated ingress/egress node pair. If a multi-hop LC is available, execution proceeds to step 506, in which the LP trunk is mapped on to the multi-hop LC and the reroute timer for the LP trunk is set. In contrast, if a multi-hop LC establishment fails, e.g., due to lack of bandwidth or available direct LCs, execution proceeds to step 508, in which a new direct LC is created, and then to step 510, in which the LP trunk is mapped on to the newly created direct LC and the reroute timer for the LP trunk is set.
FIG. 6 depicts a flowchart illustrating a method of monitoring the utilization of each established LC of an optical network in accordance with one embodiment. The method illustrated in FIG. 6 is performed at each ingress node for each established LC. Execution begins in step 600, responsive to expiration of a utilization check timer for the LC, which defines the periodicity with which each LC is checked for utilization. In step 600, a determination is made whether utilization of the LC is above an upper threshold, indicating that the LC is over-utilized, and a monitoring flag is not set. If so, execution proceeds to step 602, in which the monitoring flag is set, and then to step 604, in which an appropriate LSA is sent to all of the edge nodes advising of the fact that the utilization of the LC has exceeded the upper threshold. Upon receipt of the LSA sent in step 604, the edge nodes update their respective Preferred Hops tables appropriately. Specifically, in each table, the entry corresponding to the LC is decremented by one.
 If a negative determination is made in step 600, execution proceeds to step 605, in which a determination is made whether utilization of the LC is above an upper threshold, indicating that the LC is over-utilized, and a monitoring flag is set. If not, execution proceeds to step 606, in which a determination is made whether the utilization is between upper and lower thresholds and the monitoring flag is set.
 If both conditions are met in step 606, indicating that the utilization of the LC had previously exceeded the upper threshold (as indicated by the fact that the monitoring flag is set), but has fallen back below the upper threshold, but not below the lower threshold, execution proceeds to step 608, in which the monitoring flag is unset, and then to step 609. In step 609, an appropriate LSA is sent to all of the edge nodes advising of the fact that the utilization of the LC has returned to within acceptable limits. Upon receipt of the LSA sent in step 609, the edge nodes update their respective Preferred Hops tables appropriately. Specifically, at each table, the entry corresponding to the LC is incremented by one. It should be noted that the utilization can be averaged over longer intervals so that oscillations around the upper threshold can be reduced.
 If a negative determination is made in step 606, execution proceeds to step 610. In step 610, a determination is made whether utilization is below a lower threshold, indicating that the LC is underutilized. If so, execution proceeds to step 612. In step 612, a determination is made whether a time-to-live (“TTL”) timer has been started. The TTL timer indicates the remaining time an LC can remain underutilized before it is torn down. If in step 612 it is determined that the TTL timer has not been started, execution proceeds to step 614, in which the timer is started, and then to step 615, in which an appropriate LSA is sent to all of the edge nodes advising of the fact that the TTL timer for the LC has been started. Upon receipt of the LSA sent in step 615, the edge nodes attempt to reroute their traffic trunks on to other LCs before the TTL timer expires. If a positive determination is made in step 612, execution proceeds to step 616.
 In step 616, a determination is made whether the TTL timer has expired. If so, execution proceeds to step 618, in which the LC is torn down, and then to step 619, in which an appropriate LSA is sent to all of the edge nodes advising of the fact that the LC has been torn down. Upon receipt of the LSA sent in step 619, the edge nodes update their respective Preferred Hops tables appropriately. Specifically, at each table, the entry corresponding to the LC is decremented by one. Execution ends in step 620.
 Subsequent to execution of any one of steps 604, 609, or 615, or responsive to a negative determination in steps 610 or 616 or a positive determination in step 605, execution proceeds to step 622, in which the utilization check timer is reset, and then awaits expiration of the utilization check timer before returning to step 600.
FIG. 7 depicts the results of a simple simulation performed using an arrangement equivalent to the network 100 as an evaluation of the RITE method described herein. For simulation purposes, each of the links 106 a-106 c, 108 a-108 c is a Wavelength Division Multiplexed (“WDM”) link with eight wavelengths. The simulation has been implemented with a network simulator known as ns-2, version beta-8, by extending the existing LDP to carry optical resources (probe packets) information for LC set-up using First Fit (available wavelength along the path OXCs), teardown and LSP mapping of traffic trunks. HP/LP requests and their lifetimes are generated according to exponential capacity (mean parameter varied). Each of the LCs can accommodate 15 discrete HP/LP requests (capacity/granularity of BW requests identical). The upper and lower thresholds are respectively selected as 80% and 20% of an LC bandwidth. The reroute timer is set to five time units, which is approximately 30% of the average LC lifetimes, and the utilization timer is set to ten time units. “Widest Path” (with largest available bandwidth per wavelength) is employed for finding explicit paths.
FIG. 7 illustrates the effective routing stability provided for HP traffic, while the rerouting frequency for LP traffic is bounded with in 20% during their lifetimes, which are exponentially distributed. When the utilization of an LC is low, thereby triggering the teardown of the LC, some of the HP traffic is also rerouted and mapped on to multi-hop LSPs. For the LP traffic, as the utilization factor increases, the probability of rerouting the LP trunks to make way for HP trunks of direct LCs increases. This demonstrates the feasibility of achieving routing stability under the abovementioned criteria and, as a result, improving QoS measures such as throughput, delay, and jitter and minimizing the service signal disruption for the ongoing sessions.
 In general, if the distinction (HP vs. LP) between the classes is ignored, the reroute timer mechanism can still serve as a simple but efficient policy to provide routing stability. Constraints such as propagation delay and queuing delay for multi-hop LSPs can augment the path selection process. Traffic prediction can be taken into account while attempting to establish new LCs. A new class of traffic can be introduced such that an LC that is established and on to which traffic of such class is mapped will not be torn down even in the case of poor utilization until the LC is left without any class of traffic mapped thereon.
 Based upon the foregoing Detailed Description, it should be readily apparent that the present invention advantageously provides a simple method of providing integrated TE resource management (both bandwidth and wavelengths) and provides the routing stability to the desired classes of traffic. Further, service differentiation can be introduced by having different reroute timers for each class of traffic.
 It is believed that the operation and construction of the present invention will be apparent from the foregoing Detailed Description. While the exemplary embodiments of the invention shown and described have been characterized as being preferred, it should be readily understood that various changes and modifications could be made therein without departing from the scope of the present invention as set forth in the following claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US2151733||May 4, 1936||Mar 28, 1939||American Box Board Co||Container|
|CH283612A *||Title not available|
|FR1392029A *||Title not available|
|FR2166276A1 *||Title not available|
|GB533718A||Title not available|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7266296||Jun 11, 2003||Sep 4, 2007||Intel Corporation||Architecture and method for framing control and data bursts over 10 Gbit Ethernet with and without WAN interface sublayer support|
|US7298973||Apr 16, 2003||Nov 20, 2007||Intel Corporation||Architecture, method and system of multiple high-speed servers to network in WDM based photonic burst-switched networks|
|US7310480||Jun 18, 2003||Dec 18, 2007||Intel Corporation||Adaptive framework for closed-loop protocols over photonic burst switched networks|
|US7315693||Oct 22, 2003||Jan 1, 2008||Intel Corporation||Dynamic route discovery for optical switched networks|
|US7340169||Nov 13, 2003||Mar 4, 2008||Intel Corporation||Dynamic route discovery for optical switched networks using peer routing|
|US7359386 *||Dec 2, 2003||Apr 15, 2008||Cisco Technology, Inc.||Soft preemption feedback|
|US7428383||Feb 28, 2003||Sep 23, 2008||Intel Corporation||Architecture, method and system of WDM-based photonic burst switched networks|
|US7512130||Feb 22, 2005||Mar 31, 2009||Oki Electric Industry Co., Ltd.||Optical communication network system|
|US7526202||May 19, 2003||Apr 28, 2009||Intel Corporation||Architecture and method for framing optical control and data bursts within optical transport unit structures in photonic burst-switched networks|
|US7529257 *||Apr 30, 2003||May 5, 2009||Cisco Technology, Inc.||Method for supporting a GMPLS hierarchy through multiple routing instances|
|US7688732||Dec 23, 2003||Mar 30, 2010||Telecom Italia S.P.A.||System and method for the automatic setup of switched circuits based on traffic prediction in a telecommunications network|
|US7734176||Dec 22, 2003||Jun 8, 2010||Intel Corporation||Hybrid optical burst switching with fixed time slot architecture|
|US7848649||Feb 28, 2003||Dec 7, 2010||Intel Corporation||Method and system to frame and format optical control and data bursts in WDM-based photonic burst switched networks|
|US7907526||May 26, 2006||Mar 15, 2011||Telefonaktiebolaget L M Ericsson (Publ)||Traffic-triggered setup of label switched paths|
|US7978713 *||Mar 24, 2004||Jul 12, 2011||Nippon Telegraph And Telephone Corporation||GMPLS+IP/MPLS node and IP/MPLS node|
|US7995569||Jan 23, 2006||Aug 9, 2011||Nortel Networks Limited||Virtual routers for GMPLS networks|
|US8660427||Sep 13, 2002||Feb 25, 2014||Intel Corporation||Method and apparatus of the architecture and operation of control processing unit in wavelenght-division-multiplexed photonic burst-switched networks|
|US8948048 *||Dec 15, 2010||Feb 3, 2015||At&T Intellectual Property I, L.P.||Method and apparatus for characterizing infrastructure of a cellular network|
|US20040170165 *||Feb 28, 2003||Sep 2, 2004||Christian Maciocco||Method and system to frame and format optical control and data bursts in WDM-based photonic burst switched networks|
|US20040170431 *||Feb 28, 2003||Sep 2, 2004||Christian Maciocco||Architecture, method and system of WDM-based photonic burst switched networks|
|US20040208171 *||Apr 16, 2003||Oct 21, 2004||Shlomo Ovadia||Architecture, method and system of multiple high-speed servers to network in WDM based photonic burst-switched networks|
|US20040208172 *||Apr 17, 2003||Oct 21, 2004||Shlomo Ovadia||Modular reconfigurable multi-server system and method for high-speed networking within photonic burst-switched network|
|US20040234263 *||May 19, 2003||Nov 25, 2004||Shlomo Ovadia||Architecture and method for framing optical control and data bursts within optical transport unit structures in photonic burst-switched networks|
|US20040252995 *||Jun 11, 2003||Dec 16, 2004||Shlomo Ovadia||Architecture and method for framing control and data bursts over 10 GBIT Ethernet with and without WAN interface sublayer support|
|US20040258407 *||Jun 18, 2003||Dec 23, 2004||Christian Maciocco||Adaptive framework for closed-loop protocols over photonic burst switched networks|
|US20050030951 *||Aug 6, 2003||Feb 10, 2005||Christian Maciocco||Reservation protocol signaling extensions for optical switched networks|
|US20050068968 *||Sep 30, 2003||Mar 31, 2005||Shlomo Ovadia||Optical-switched (OS) network to OS network routing using extended border gateway protocol|
|US20050089327 *||Oct 22, 2003||Apr 28, 2005||Shlomo Ovadia||Dynamic route discovery for optical switched networks|
|US20050105905 *||Nov 13, 2003||May 19, 2005||Shlomo Ovadia||Dynamic route discovery for optical switched networks using peer routing|
|US20050117512 *||Dec 2, 2003||Jun 2, 2005||Cisco Technology, Inc.||Soft preemption feedback|
|US20050135806 *||Dec 22, 2003||Jun 23, 2005||Manav Mishra||Hybrid optical burst switching with fixed time slot architecture|
|US20050175183 *||Feb 9, 2004||Aug 11, 2005||Shlomo Ovadia||Method and architecture for secure transmission of data within optical switched networks|
|US20050177749 *||Mar 18, 2004||Aug 11, 2005||Shlomo Ovadia||Method and architecture for security key generation and distribution within optical switched networks|
|US20050185586 *||Feb 22, 2005||Aug 25, 2005||Oki Electric Industry Co., Ltd.||Optical communication network system|
|US20060018313 *||Mar 24, 2004||Jan 26, 2006||Nippon Telegraph And Telephone Corporation||Gmplsmpls node and ip mpls node|
|US20120155319 *||Dec 15, 2010||Jun 21, 2012||Alexandre Gerber||Method and apparatus for characterizing infrastructure of a cellular network|
|EP1695313A2 *||Nov 10, 2004||Aug 30, 2006||Cisco Technology, Inc.||Soft preemption feedback|
|WO2005064983A1 *||Dec 23, 2003||Jul 14, 2005||Telecom Italia Spa||System and method for the automatic setup of switched circuits based on traffic prediction in a telecommunications network|
|WO2007138443A1 *||May 25, 2007||Dec 6, 2007||Ericsson Telefon Ab L M||Traffic-triggered setup of label switched paths|
|U.S. Classification||370/237, 370/351, 370/400|
|International Classification||H04L12/56, H04J14/02|
|Cooperative Classification||H04J14/0241, H04L45/50, H04L47/10, H04J14/0287, H04L45/00, H04J14/0284, H04J14/0227|
|European Classification||H04J14/02P, H04L45/50, H04L45/00, H04L47/10, H04J14/02M|
|Jun 24, 2002||AS||Assignment|
Owner name: ALCATEL, SOCIETE ANONYME, FRANCE
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:CHIRUVOLU, GIRISH;REEL/FRAME:013059/0954
Effective date: 20020612
|Sep 18, 2007||AS||Assignment|
Owner name: ALCATEL LUCENT, FRANCE
Free format text: CHANGE OF NAME;ASSIGNOR:ALCATEL;REEL/FRAME:019839/0098
Effective date: 20061130
|Apr 26, 2011||FPAY||Fee payment|
Year of fee payment: 4
|Jan 30, 2013||AS||Assignment|
Owner name: CREDIT SUISSE AG, NEW YORK
Free format text: SECURITY AGREEMENT;ASSIGNOR:LUCENT, ALCATEL;REEL/FRAME:029821/0001
Effective date: 20130130
Owner name: CREDIT SUISSE AG, NEW YORK
Free format text: SECURITY AGREEMENT;ASSIGNOR:ALCATEL LUCENT;REEL/FRAME:029821/0001
Effective date: 20130130
|Sep 30, 2014||AS||Assignment|
Owner name: ALCATEL LUCENT, FRANCE
Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:CREDIT SUISSE AG;REEL/FRAME:033868/0001
Effective date: 20140819
|Apr 23, 2015||FPAY||Fee payment|
Year of fee payment: 8